Add mtbseq workflow

assets/schema_input.json

@@ -12,6 +12,12 @@
                 "pattern": "^\\S+$",
                 "errorMessage": "Sample name must be provided and cannot contain spaces",
                 "meta": ["id"]
+            },
+             "library": {
+                "type": "string",
+                "pattern": "^\\S+$",
+                "errorMessage": "Library preparation protocol name must be provided and cannot contain spaces",
+                "meta": ["library"]
             },
             "fastq_1": {
                 "type": "string",

conf/mtbseq_base.config

@@ -0,0 +1,73 @@
+params {
+     mtbseq_parallel = ""
+     mtbseq_only_qc = false
+    //-------------------------------------
+    // OPTIONS FROM MTBSEQ MANUAL
+    //-------------------------------------
+
+    // Hidden parameters
+    mtbseq_user                       = "root"
+    mtbseq_path                       = "MTBseq"
+
+
+    // Name of the project and the analysis
+    mtbseq_cohort_tsv                 = null
+    mtbseq_project                    = "mtbseqnf"
+
+    // Setting this OPTION will add an additional filter that excludes all variants except SNPs.
+    mtbseq_snp_vars = false
+
+    // Setting this OPTION has major implications on how the mapping data for each position is processed. By default, the majority allele is called and taken for further calculations.
+    // If the --lowfreq_vars OPTION is set, MTBseq will consider the majority allele distinct from wild type, if such an allele is present.
+    // This means that only in this detection mode, MTBseq will report variants present only in subpopulations, i.e. low frequency mutations.
+    // Of course, OPTIONS --mincovf, --mincovr, --minphred and --minfreq need to be set accordingly.
+    // Please be aware that output generated in this detection mode should not be used for phylogenetic analysis.
+    mtbseq_lowfreq_vars = false
+
+    // The OPTION sets a threshold for the sequence data quality to be used for the mpileup creation
+    mtbseq_minbqual = 13
+
+    // The OPTION sets a minimum forward read coverage threshold. Alleles must have a forward coverage of this VALUE or higher to be considered.
+    mtbseq_mincovf = 4
+
+    // The OPTION sets a minimum reverse read coverage threshold. Alleles must have a reverse coverage of this VALUE or higher to be considered.
+    mtbseq_mincovr = 4
+
+    // The OPTION sets a minimum number of reads indicating an allele with a phred score of at least 20.
+    mtbseq_minphred = 4
+
+    // The OPTION sets a minimum frequency for an allele.
+    mtbseq_minfreq = 75
+
+    // The option sets a minimum percentage of samples with unambiguous information for position.
+    mtbseq_unambig = 95
+
+    // The OPTION sets a window size in which the algorithm scans for the occurrence of multiple variants within the same sample.
+    // If more than one variant occurs within this window in the same sample, the positions will be excluded.
+    mtbseq_window = 12
+
+    // The OPTION sets a SNP distance that is used to classify samples into groups of samples, using agglomerative clustering.
+    // If SNP distances between samples are less or equal this VALUE, they are grouped together.
+    mtbseq_distance = 12
+
+    // This OPTION sets the reference genome for the read mapping.
+    mtbseq_ref_and_indexes_path = "${projectDir}/data/mtbseq-references/ref/"
+
+    // This OPTION sets a list of known variant positions associated to drug resistance for resistance prediction.
+    mtbseq_resilist = "${projectDir}/data/mtbseq-references/res/MTB_Resistance_Mediating.txt"
+
+    // This OPTION sets a list of interesting regions to be used for annotation of detected variants
+    mtbseq_intregions = "${projectDir}/data/mtbseq-references/res/MTB_Extended_Resistance_Mediating.txt"
+
+    // This OPTION specifies a gene categories file to annotate essential and non-essential genes as well as repetitive regions. SNPs in repetitive regions will be excluded for phylogenetic analysis.
+    mtbseq_categories = "${projectDir}/data/mtbseq-references/cat/MTB_Gene_Categories.txt"
+
+    // This OPTION specifies a file for base quality recalibration. The list must be in VCF format and should contain known SNPs.
+    mtbseq_basecalib = "${projectDir}/data/mtbseq-references/res/MTB_Base_Calibration_List.vcf"
+
+
+
+
+}
+
+

conf/mtbseq_parallel_test.config

@@ -0,0 +1,28 @@
+process {
+    resourceLimits = [
+        cpus: 4,
+        memory: '8.GB',
+        time: '3.h'
+    ]
+}
+
+
+params {
+    config_profile_name        = 'MTBseq mode using parallel processing test profile'
+    config_profile_description = 'Minimal test to check if MTBseq mode with parallel processing is working as intended'
+
+
+    input  = "${projectDir}/data/test_data/test_3_samples.csv"
+
+    mtbseq_cohort_tsv  = "${projectDir}/data/test_data/test_3_cohort.tsv"
+
+    outdir = "test-profile-results"
+
+    mode = "mtbseq"
+
+    mtbseq_parallel = true
+
+
+}
+
+

conf/mtbseq_test.config

@@ -0,0 +1,25 @@
+process {
+    resourceLimits = [
+        cpus: 4,
+        memory: '8.GB',
+        time: '3.h'
+    ]
+}
+
+
+params {
+    config_profile_name        = 'MTBseq Test profile'
+    config_profile_description = 'Minimal test to check if MTBseq mode is working as intended'
+
+    input  = "${projectDir}/data/test_data/test_3_samples.csv"
+
+    mtbseq_cohort_tsv  = "${projectDir}/data/test_data/test_3_cohort.tsv"
+
+    outdir = "mtbseq-test-profile-results"
+
+    mode = "mtbseq"
+
+
+}
+
+

data/mtbseq-references/Readme.md

@@ -0,0 +1,7 @@
+## NOTE ABOUT THE REFERENCE FILES
+
+These references are copied from `v1.0.3` of [MTBseq source tree](https://github.com/ngs-fzb/MTBseq_source/tree/v1.0.3/var).
+
+
+These are copied here to be able to stage this files directly from the `projectDir` rather than depending upon the packaging option for `MTBseq` such as `conda` or `docker` container.
+